Fructose and pancreatic cancer

ResearchBlogging.orgI hate science press releases.

Well, not exactly. I hate science press releases that hype a study beyond its importance. I hate it even more when the investigators who published the study make statements not justified by the study and use the study as a jumping off point to speculate wildly. True, it’s not always the fault of the investigators, particularly if they don’t have much experience dealing with the press, but all too often scientists fall prey to the tendency to gab glibly and give the reporter what he or she wants: Pithy, juicy quotes that relate the results to what the reporter wants them related to. It’s irritating as hell, not so much because it’s pure self-promotion. (After all, self-promotion is not in and of itself a bad thing) but rather because it’s almost inevitably an excuse for the investigators to say what they want without peer pesky peer reviewers telling them that they should keep their remarks focused on what the evidence will support. Often these press releases lead to credulous news stories that make conclusions that aren’t justified from the actual study. Sometimes an investigators’ comments are taken out of context. Sometimes the investigator says something dumb. Sometimes it’s all three.

There’s a certain Reuters story entitled Cancer cells slurp up fructose, U.S. study finds making the rounds, and it’s being represented as yet more evidence about the evils of high fructose corn syrup. That it might be viewed as a few years in the future, after followup studies have been done, but for right now all it is is an intriguing study being used to serve an agenda that it doesn’t serve well:

WASHINGTON–Pancreatic tumour cells use fructose to divide and proliferate, U.S. researchers said Monday in a study that challenges the common wisdom that all sugars are the same.

Tumour cells fed both glucose and fructose used the two sugars in two different ways, the team at the University of California Los Angeles found.

They said their finding, published in the journal Cancer Research, may help explain other studies that have linked fructose intake with pancreatic cancer, one of the deadliest cancer types.

“These findings show that cancer cells can readily metabolize fructose to increase proliferation,” Dr. Anthony Heaney of UCLA’s Jonsson Cancer Center and colleagues wrote.

“They have major significance for cancer patients given dietary refined fructose consumption, and indicate that efforts to reduce refined fructose intake or inhibit fructose-mediated actions may disrupt cancer growth.”

Just from this description it actually sounds like a rather interesting study, but right away I don’t like the implications being ascribed to its results. They’re way premature. You’ll see what I mean in a minute. First, let’s see what Dr. Heaney says:

The industry has also argued that sugar is sugar.

Heaney said his team found otherwise. They grew pancreatic cancer cells in lab dishes and fed them both glucose and fructose.

Tumour cells thrive on sugar but they used the fructose to proliferate. “Importantly, fructose and glucose metabolism are quite different,” Heaney’s team wrote.

“I think this paper has a lot of public health implications. Hopefully, at the federal level there will be some effort to step back on the amount of high fructose corn syrup in our diets,” Heaney said in a statement.

Oh, nooooo! Bad move. Way premature.

Dr. Heany appears to be implying that high fructose corn syrum causes pancreatic cancer or somehow makes it worse and that this paper has major public health implications when it comes to cancer. It does not, at least not yet. It’s possible the findings that spring out of subsequent studies spawned by this paper might have public health implications, but this paper alone is not enough to make such a bold statement. Fortunately, Heany at least has some interesting and potentially important observations to report. Unfortunately, as I guessed when I read just the news story, this study only looks at pancreatic cells in culture. On that basis alone, I would chastise Dr. Heaney for making such policy statements. There may or may not be legitimate science-based reasons to regulate the content of high fructose corn syrup in food and beverages based on their health effects, but this is nothing more than a study in which the effects of fructose were compared to glucose on pancreatic cancer cell metabolism. He didn’t even include animal studies to show something like, for example, that a high fructose diet in mice would accelerate the growth of pancreatic cancer. There’s nothing of the sort in this paper. It’s all cell culture. So while it’s an interesting study, it is not particularly strong evidence that high fructose corn syrup leads to pancreatic cancer.

Because it’s a scientifically intriguing study, even in spite of my annoyance with the way it was reported, I thought I’d go straight to the source and look up the actual article in Cancer Research, entitled Fructose Induces Transketolase Flux to Promote Pancreatic Cancer Growth. Before I get into the discussion of the paper, I thin it’s worth reiterating that common table sugar is sucrose, which is a disaccharide consisting of one fructose molecule and one glucose molecule bound together, or, to put it more simply, a larger sugar made of two simple sugars linked together by a chemical bond with the following chemical structure:


The fructose molecule is the one consisting of a ring of four carbons and an oxygen, forming a pentagram. (Appropriate, no, given how it’s been painted as the root of all evil?) Glucose, on the other hand, although it has the same number of carbon atoms has a six-sided ring consisting of five carbons and an oxygen. In any case, any time you eat table sugar, after digestion of the disaccharide by an enzyme known as sucrase secreted by the villi lining the intestines, you’re getting an equal mixture (in terms of moles) of fructose and glucose. In the abstract, the authors point out that traditionally glucose and fructose have been considered interchangeable simple sugars that are metabolized similarly. However, anyone who’s taken basic biochemistry knows that there are differences in how fructose and glucose are metabolized. Here’s a nice simple diagram cribbed from Wikipedia that shows the difference:


I’m having acid flashbacks to graduate level biochemistry class.

Of course, I shouldn’t, at least not due to this simplistic diagram. Back in biochemistry, I had to memorize not just the chemical structure of each intermediary in glycolysis, but its chirality (handedness) and the enzymes and cofactors involved. This is much simpler (simple enough for medical school). This is nothing (or would have been nothing 25 years ago). In any case, as you can see, fructose can be metabolized the same as glucose beginning at the fructose 6-phosphate step. Even so, fructose tends to be metabolized differently than glucose. Perhaps the most important difference is that fructose tends to be metabolized in the liver, while glucose is metabolized in every cell in the body. That’s because in the liver an enzyme known as fructokinase adds a phosphate group to fructose and is then metabolized through a process known as fructolysis. Although glycolysis (the metabolic process by which glucose is metabolized initially) and fructolysis share some intermediate molecules, the metabolic fates are different. Fructose tends to be used in the liver to make glycogen, which is the form in which the liver stores energy for short term use. It can also be used to produce the building blocks of fatty acids.

Perhaps the most important uses of fructose in metabolism, though, are in the synthesis of amino acids for DNA synthesis. The authors note:

The nonoxidative pentose phosphate pathway (PPP), which allows six-carbon glucose conversion to five-carbon ribose for DNA or RNA synthesis, is of utmost importance for the proliferation process and produces >85% of the ribose recovered from tumor nucleic acids (2). The nonoxidative PPP, controlled by transketolase (TK) enzyme reactions, is encoded by three human TK genes: TKT, TKTL1, and TKTL2 (3, 4). Clinically, patients with extensive cancer burden have a tendency to develop thiamine depletion, which is a cofactor for TK-mediated reactions, emphasizing the importance of the nonoxidative PPP for tumor growth.

The authors also cite two prospective studies that found an increased relative risk of pancreatic cancer associated with the consumption of fructose, although I must admit that I’m not all that impressed with them. For one thing, one study showed fairly weak associations (relative risk less than 2.0, with one of the trends not being statistically significant and another being barely significant). The second study was only slighly more convincing. In any case, let’s just say that the evidence linking fructose intake to pancreatic cancer, while somewhat suggestive, is by no means bulletproof, and it’s not clear that either of these studies adequately controlled for confounding factors. I’d say it’s probably worth looking into.

So look into it the Heany’s group did. The first finding was one that didn’t particularly surprise me, namely that pancreatic cancer cells could use fructose as a substrate and grow just as fast as they could in glucose. They used multiple assays to measure proliferation and found that fructose and glucose produced the same rate of proliferation. Finally, they tested cells isolated from freshly resected human pancreatic cancer specimens and found…the same thing! Basically, they showed that, although glucose is the preferred fuel for pancreatic tumor growth, fructose could serve just as well.

Next, Heaney’s team did a bunch of metabolomic studies, looking at the products of glucose and fructose metabolism in order to determine. Basically, they incubated cancer cells in medium containing either glucose or fructose labeled with the isotope of carbon, 13C, which allowed them to track the products of metabolism using mass spectroscopy. What they found was that, although glucose and fructose can share the same metabolic pathway after fructose 6-phosphate, in the cell glucose and fructose are metabolized differently. Glucose, not unexpectedly, was used mostly for glycolysis:

The results derived from the Panc-1 pancreatic cancer cells are presented, and as expected, a large proportion of the 13C-labeled glucose administered to the pancreatic cancer cells entered glycolysis and was metabolized to generate lactate (Fig. 2A) and CO2 (Fig. 2B). In contrast, a comparatively small fraction of 13C-labeled fructose was metabolized to generate lactate and CO2, resulting in 800% lower lactate and 350% lower CO2 production for fructose in comparison with glucose. Likewise, a significant proportion of glucose metabolism contributed to fatty acid synthesis as evidenced by glucose-derived 13C-labeled behenic acid (Fig. 2C, light bar) and palmitate (Fig. 2D, light bar), but 150% lower levels of behenic acid (Fig. 2C, dark bar) and palmitate levels (Fig. 2D, dark bar) were derived from 13C-labeled fructose (similar results were seen for all C16 to C26 fatty acids).

So how was fructose metabolized in pancreatic cancer cells? For the most part, it was used to generate nucleic acid synthesis. Compared to glucose, fructose induces is preferentially metabolized via the nonoxidative pentose phosphate pathway to synthesize nucleic acids and increase uric acid production. What this means is that fructose provides the raw materials for cancer cells to make more DNA, which cells must do in order to divide and proliferate. The differences are summarized in the following chart:


Panel A represents glucose metabolism and the relative proportions of each metabolic process glucose is shunted to, while panel B shows fructose. Compared to glucose, fructose enters the nonoxidative pentose phosphate shunt and is used to synthesize nucleic acids.

It’s well known that cancer cells suck up and metabolize glucose far more avidly than normal cells. The reason is that cancer cells, even when there is oxygen present, still tend to use anaerobic metabolism, which produces far less energy per molecule of glcuose and thus requiring a lot mroe glucose. This behavior is known as the Warburg effect and was first described by Otto Warburg in 1928. That’s the basis for a drug that I’ve written about pretty extensively over the last couple of years, namely dichloroacetate, which selectively targets the Warburg effect. Heaney’s results are potentially important in that they suggest that not just aberrant glucose metabolism but aberrant fructose metabolism might be involved in the pathogenesis of pancreatic cancer. Whether diet contributes to these changes or not is an open question. Unfortunately, Heaney takes his potentially important results, and yokes them to a policy using far stronger terms than are justified:

“The bottom line is the modern diet contains a lot of refined sugar including fructose and it’s a hidden danger implicated in a lot of modern diseases, such as obesity, diabetes and fatty liver,” said Heaney, who also serves as director of the Pituitary Tumor and Neuroendocrine Program at UCLA. “In this study, we show that cancers can use fructose just as readily as glucose to fuel their growth.”


As in anti-smoking campaigns, a federal effort should be launched to reduce refined fructose intake, Heaney said.

“I think this paper has a lot of public health implications,” Heaney said. “Hopefully, at the federal level there will be some effort to step back on the amount of HFCS in our diets.”

And that, is how you go from cell culture work to sweeping policy recommendations. Pass Go, collect $200. Rock and roll.

As I mentioned before, there very well may be legitimate, science-based reasons to be worried about fructose and high fructose corn syrup. Steve Novella recently wrote a good review of the topic in which he wanted to point out that glucose and fructose are sugars, as is the main component of high fructose corn syrup. It’s a high calorie, highly processed substance that has no other nutritional value other than as pure fuel. What Heaney’s study doesn’t really answer is just how different from the case in normal pancreatic cells is fructose metabolism in pancreatic cells or if it’s even different? In the current study, HPDE6 cells, which are a cell line derived from normal pancreas cells, behaved the same as all the pancreatic cancer cell lines. Fructose supported its growth and was used primarily for ribose and nucleic acid synthesis.

There’s a lot of hard work that needs to be done to validate the cell culture findings reported by Heaney’s group, not the least of which is testing it in animal models. It’s far too early to make any sort of recommendations about high fructose corn syrup and diet based on this study, and Heaney should have known better. Next up, I can see an article on touting this study as “proof” that high fructose corn syrup is pure, concentrated evil and that food manufacturers are out to kill us all, and Heaney will have given Mike Adams some choice quotes to use for that article.


Liu, H., Huang, D., McArthur, D., Boros, L., Nissen, N., & Heaney, A. (2010). Fructose Induces Transketolase Flux to Promote Pancreatic Cancer Growth Cancer Research, 70 (15), 6368-6376 DOI: 10.1158/0008-5472.CAN-09-4615